learning_rate: 0.0002
train_batch_size: 1
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
training_steps: 600
mixed_precision_training: Native AMP

Training results

Framework versions

PEFT 0.15.2
Transformers 4.52.4
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.2

How to use this model

This model is a LoRA fine-tuned adapter for EFU General Insurance Q&A.
It must be loaded on top of the base LLaMA model.

python

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

# Base model
base_model = "meta-llama/Llama-2-7b-hf"

# LoRA adapter (this repo)
adapter_model = "DamnOnic/efu-general-demo"

# Load tokenizer and base model
tokenizer = AutoTokenizer.from_pretrained(base_model)
base = AutoModelForCausalLM.from_pretrained(base_model, device_map="auto", load_in_8bit=True)

# Load fine-tuned adapter
model = PeftModel.from_pretrained(base, adapter_model)

# Ask questions
def ask(question, max_new_tokens=200):
    prompt = f"### Instruction: {question}\n\n### Input:\n\n### Response:"
    inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
    outputs = model.generate(**inputs, max_new_tokens=max_new_tokens, temperature=0.7, top_p=0.9)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

print(ask("What is EFU General Insurance?"))